_lambdar.yml file can be used to customise your container image. Lambdar tries to populate as much information as it can, but if you need to override any options you’re welcome to do so. You can also write your own Dockerfile and use lambdar to build it if you prefer.
Lambdar uses these when it talks to AWS. This information is needed to:
docker pushand upload your container.
If you don’t supply this information you can still build a container if you have Docker installed, or build a Dockerfile if you don’t.
This is used to name (via
docker tag) your container when it’s built. It also forms part of the ECR repository URL.
This tells lambdar what version of R to install in your container. Lambdar pre-fills this with your current R version, but you can change it if you need to. Make sure you specify all three digits of the version number, e.g.
Any packages listed here will be installed when the container is built. Lambdar will try and determine which packages you need using
renv::dependencies() and will auto-fill this list.
If you need any extra system libraries installing, mention them here. The base AWS image is based on CentOS, so packages will be installed by
Lambdar can’t work out which Linux package you need so this section is up to you.
If your code needs any specific environment variables, enter them here. At the moment
yaml lists are not supported because they don’t translate to the correct data structure in R, so copy the examples in the template.